introduction TO DISSERTATION RESEARCH
The challenge of autonomy is not replacing humans but designing systems that allow humans and machines to work together effectively.
The following projects represent a set of research investigations into HUMAN PERFORMANCE AND human autonomy teaming. Each project approaches the problem from a different perspective, examining how trust develops, how coordination emerges between humans and autonomous systems, and how these interactions influence team performance.
This work was conducted in the Cyber-Human-Physical Systems Lab at UC Davis. I had the privilege of working for the NASA HOME STRI.
Real-Time Inference and Prediction of Trust in Human-Autonomy Teaming
My research examines how trust evolves during interactions between humans and autonomous systems. As people observe system behavior, they continually update their expectations about reliability, capability, and intent. Understanding these dynamics requires objective and unobtrusive methods for inferring trust as it changes over time.
This project develops embedded and physiological measures, together with predictive models, to assess trust dynamics within individual interactions in human autonomy teams. The goal is to enable real time inference of trust states and support adaptive system behaviors that promote effective human autonomy teaming.
This work is conducted in collaboration with Dr. Allie Anderson and Dr. Torin Clark at the University of Colorado Boulder.
publications
HUMAN-ROBOT INTERACTION
Human–autonomy teams are increasingly expected to operate in demanding environments such as search and rescue, hazard response, and space exploration. In these settings, autonomous agents coordinate their actions with human operators to achieve shared goals. While autonomous technologies continue to advance, the effectiveness of these teams often depends on the human operator’s ability to apply knowledge and expertise in dynamic and uncertain conditions.
A central challenge in human autonomy teaming is developing systems that can support operators during complex tasks. One important step toward this goal is the ability to estimate changes in human cognitive state in real time.
In this project, we examined both behavioral and neurophysiological responses as participants piloted a team of robots in a target identification and acquisition task. The primary experimental factor was the difficulty of estimating the kinematic state of the robot group. In addition to gaze behavior and pilot inputs, we analyzed EEG spectral power and measures of functional connectivity to investigate how cognitive state relates to performance during human–robot team coordination.
publications
Data Driven Personalized Training of Next Generation Workforce
Humans will continue to play a central role in future manufacturing as operators, designers, and decision makers. At the same time, the systems they interact with are becoming increasingly complex with the growth of industrial cyber physical systems and the Industrial Internet of Things. This project developed a data driven approach for modeling and analyzing human manual expertise during a grinding task. The goal is to support more effective collaboration between humans and machines and to help preserve and transfer skilled knowledge across generations. This work was conducted in collaboration with Dr. Barbara Linke.
In this study, we examined how behavior and performance relate during a manual grinding task performed by participants with different skill levels. The results revealed distinct sensorimotor patterns associated with different techniques, and these techniques were shown to influence task performance.
